Quoting Ben Affleck
Ben Affleck said he has been interested in computers since childhood, and that his interest grew when filmmaking shifted from analog film to digital, according to remarks he made about his technical work.
He said the visual effects workflow has involved machine learning for many years. He described writing what he called pretty shitty Python scripts, and said he works with convolutional neural networks, which he identified as precursors to what the transformer can do.
Affleck explained that these networks process far more computation simultaneously. He described working with a tensor, which he defined as the numerical translation of a visual image into numbers, covering the batch number, the frame number, and the red, green and blue values of each pixel in each frame.
He said a convolutional neural network is used to identify patterns in that data, a process he called edge detection or feature extraction. That means identifying enough patterns to determine where something like a window ledge is, so a green screen image can be more easily removed and replaced with something else.
Why it matters: Affleck's account describes machine learning techniques, including convolutional neural networks and tensors, as established parts of the visual effects pipeline rather than new additions.
Based on reporting from the original publisher. Visit the source for full context and later updates.
Publisher excerpt
I've always been kind of into computers since I was young. And then when film started to move from analog film to digital, I became more interested in that aspect of it. And the visual effects workflow for many years has included machine learning. So I can write like pretty shitty Python scripts and stuff like that because with convolutional neural networks, which were the sort of precursors to what the transformer can do, which is just much more computation simultaneously, you would do things like look at what's c